A prediction information derivation unit derives inter prediction information candidates from inter prediction information of coded prediction blocks neighboring a prediction block subject to coding within the same picture as the prediction block subject to coding and inter prediction information of a prediction block in a coded picture that is different from the prediction block subject to coding. The prediction information derivation unit determines an inter prediction information candidate to be used for inter prediction of the prediction block subject to coding from the inter prediction information candidates that have been derived. A second bitstream generation unit codes an index that indicates the inter prediction information candidate based on the number of the inter prediction information candidates.
Legal claims defining the scope of protection. Each claim is shown in both the original legal language and a plain English translation.
1. A moving picture decoding device adapted to decode a bitstream in which moving pictures are coded in units of blocks obtained by partitioning each picture of the moving pictures, comprising: a first bitstream decoding unit configured to set the number of merge candidates; a second bitstream decoding unit configured to decode information indicating indices of the merge candidates; a spatial merge candidate generation unit configured to derive spatial merge candidates from a first predefined number of blocks neighboring a prediction block subject to decoding; a temporal merge candidate generation unit configured to derive a temporal merge candidate from a block that exists at the same position as or near a prediction block subject to decoding in a decoded picture that is different from the prediction block subject to decoding; a merge candidate addition unit configured to add the spatial merge candidates and the temporal merge candidates to a merge candidate list; a coding information selection unit configured to select a merge candidate from the merge candidates added to the merge list; and a motion compensation prediction unit configured to perform inter prediction of the prediction block subject to decoding by the merge candidate thus selected, wherein the second bitstream decoding unit derives the indices of the merge candidates based on the number of the merge candidates; and the spatial merge candidate generation unit stops deriving the spatial merge candidates when the number of the derived spatial merge candidates reaches a second predefined number smaller than the first predefined number.
A video decoding device decodes a bitstream of coded moving pictures divided into blocks. It initializes a merge candidate list size, then decodes indices pointing to merge candidates within that list. Spatial merge candidates are derived from a predefined number of neighboring blocks of the block being decoded. A temporal merge candidate is derived from a block in a previously decoded picture, located at or near the same position as the current block. The spatial and temporal candidates are added to the merge candidate list. A merge candidate is selected from the list, and this candidate is used to perform inter prediction for the current block. Critically, index decoding is based on the current merge candidate list size. Spatial candidate derivation stops when a second, smaller predefined number of candidates is reached, optimizing the search.
2. A moving picture decoding method adapted to decode a bitstream in which moving pictures are coded in units of blocks obtained by partitioning each picture of the moving pictures, comprising: a first bitstream decoding step of setting the number of merge candidates; a second bitstream decoding step of decoding information indicating indices of the merge candidates; a spatial merge candidate generation step of deriving spatial merge candidates from a first predefined number of blocks neighboring a prediction block subject to decoding; a temporal merge candidate generation step of deriving a temporal merge candidate from a block that exists at the same position as or near a prediction block subject to decoding in a decoded picture that is different from the prediction block subject to decoding; a merge candidate addition step of adding the spatial merge candidates and the temporal merge candidates to a merge candidate list; a coding information selection step of selecting a merge candidate from the merge candidates added to the merge list; and a motion compensation prediction step of performing inter prediction of the prediction block subject to decoding by the merge candidate thus selected, wherein the second bitstream decoding step derives the indices of the merge candidates based on the number of the merge candidates; and the spatial merge candidate generation step stops deriving spatial merge candidates when the number of the derived spatial merge candidates reaches a second predefined number smaller than the first predefined number.
A video decoding method decodes a bitstream of coded moving pictures divided into blocks. The process involves initializing the size of a merge candidate list, followed by decoding indices that identify merge candidates in the list. Spatial merge candidates are obtained from a set of neighboring blocks. A temporal merge candidate is derived from a co-located or nearby block in a different, previously decoded picture. These spatial and temporal candidates are added to the merge candidate list. Then, a merge candidate is selected, enabling inter prediction for the current block. Index decoding adjusts based on the current merge candidate list size. To optimize performance, spatial candidate derivation is halted once a limited number of candidates is achieved, where that limit is smaller than the original number of neighboring blocks.
3. A non-transitory computer-readable recording medium having embedded thereon a moving picture decoding program adapted to decode a bitstream in which moving pictures are coded in units of blocks obtained by partitioning each picture of the moving pictures, the moving picture decoding program comprising: a first bitstream decoding step of setting the number of merge candidates; a second bitstream decoding step of decoding information indicating indices of the merge candidates; a spatial merge candidate generation step of deriving spatial merge candidates from a first predefined number of blocks neighboring a prediction block subject to decoding; a temporal merge candidate generation step of deriving a temporal merge candidate from a block that exists at the same position as or near a prediction block subject to decoding in a decoded picture that is different from the prediction block subject to decoding; a merge candidate addition step of adding the spatial merge candidates and the temporal merge candidates to a merge candidate list; a coding information selection step of selecting a merge candidate from the merge candidates added to the merge list; and a motion compensation prediction step of performing inter prediction of the prediction block subject to decoding by the merge candidate thus selected, wherein the second bitstream decoding step derives the indices of the merge candidates based on the number of the merge candidates; and the spatial merge candidate generation step stops deriving the spatial merge candidates when the number of the derived spatial merge candidates reaches a second predefined number smaller than the first predefined number.
A non-transitory computer-readable medium stores a video decoding program that decodes a bitstream of coded moving pictures divided into blocks. The program initializes a merge candidate list size and then decodes indices specifying merge candidates within this list. The program derives spatial merge candidates from neighboring blocks. A temporal merge candidate is derived from a block in a decoded picture at or near the same position as the current block. The spatial and temporal candidates are added to the merge candidate list. A merge candidate is selected for inter prediction. Index decoding relies on the merge candidate list size. The program halts spatial candidate derivation when a set number of spatial candidates has been found, where this limit is lower than the total initial neighboring blocks.
Cooperative Patent Classification codes for this invention. Click any code to explore related patents in that topic.
April 30, 2014
June 20, 2017
Browse 5M+ US patents with plain-English claim translations and AI-generated analysis.